Splunk Search

How do I write this SQL query as a join search in Splunk?

pm771
Communicator

I will ask my question using online forum as an example.

It has Event Log that tracks all user actions from login to logout. Contains the following significant fields: EventTime, UserID, IP address, Action.

We want to write a search that identifies users that were banned ( Action='Ban' ) and then quickly came back under another UserID, but still with the same IP.

In SQL I would use the following SELF JOIN (assuming TimeDiff is a valid function returning seconds):

SELECT ev1.IP,  ev1.UserID, ev1.EventTime, ev2.UserID, ev2.EventTime, ev2.Action
FROM EventLog ev1
JOIN Eventlog ev2 ON ev1.IP = ev2.IP AND 
                     ev1.UserID  != ev2.UserID AND
                     ev1.EventTime  <  ev2.EventTime AND
                     TimeDiff(ev2.EventTime, ev1.EventTime) <= 60
WHERE ev1.Action = 'Ban'

How to write such a search in Splunk?

0 Karma

sundareshr
Legend

Try this

index=* [search index=* action=* | dedup IP | table IP] | eventstats count by IP | where count>1 | streamstats window=1 current=f latest(_time) as nextattempt by IP | eval time_diff=abs(nextattempt-_time) | where time_diff<=60
0 Karma

pm771
Communicator

I had an unfortunate typo in my SQL query. Second line of ON condition was incorrect, should've been ev1.UserID != ev2.UserID.

Now that it's fixed, could you please re-check your answer?

TIA

0 Karma
Get Updates on the Splunk Community!

Extending Observability Content to Splunk Cloud

Watch Now!   In this Extending Observability Content to Splunk Cloud Tech Talk, you'll see how to leverage ...

More Control Over Your Monitoring Costs with Archived Metrics!

What if there was a way you could keep all the metrics data you need while saving on storage costs?This is now ...

New in Observability Cloud - Explicit Bucket Histograms

Splunk introduces native support for histograms as a metric data type within Observability Cloud with Explicit ...